Our long term goal is to understand the structure, function and development of cat retina. This tissue is part of the central nervous system, yet it is relatively simple, containing ony about 60 types of neuron. We have shown by computer reconstruction from electron micrographs of serial sections that specific types are wired with extraordinary precision into may parallel circuits. This anatomical knowledge, plus information regarding the neurochemistry and physiology of certain cell types, has led to specific hypotheses regarding function. Thus, models have emerged to explain the receptive field properties of on gaglion cells (both X and Y), how these properties are maintained in day and night vision, and how the sensitivity of the circuits is adjusted to rapid changes in light intensity (gain control) and to slow changes (dark adaptation). We now propose: (1) Extend knowledge of structure by determining additional important features of the wiring: (a) identify rod and cone bipolar circuits to off ganglion cells. (b) identify amacrine circuits to on ganglion cells (c) identify input/output circuits of putative control elements, the GABA- ergic interplexiform cell and the dopaminergic amacrine cell. The method will combine cell identification by serial EM reconstruction. (II) Test theories of function that have emerged from our structural studies: (a) establish an in vitro preparation for physiological recording. (b) use this preparation to test the hypothesized roles of GABA and the interplexiform cell in gain control. (c) test the hypothesized contributions of types A and B horizontal cells to gain control and to the ganglion cell receptive fields. (III) Detemine the normal development of identified circuits. These circuits, absent at birth, develop explosively between 1-4 weeks postnatal. (a) determine whether precision wiring in adult is established without error or whether it emerges gradually from connetions that are at first imprecise. (b) determine when antigens specific for synapses in identified circuits first appear. The methods will combine immunocytochemistry (using existing monoclonal antibodies) with serial EM reconstruction. The proposed studies address fundamental questions regarding the organization and development of identified circuits in mammalian brain and are specifically releveant to organization are shared by cat and human.